Non-traditional Uses for Robotics in Packaging

Automation Experts Apply Automotive Industry Knowledge in a Novel Way

By Sam Hoff, Patti Engineering CEO and Scott Steele, President of SWS Consulting

Robots have entered the mainstream of package production covering many routine material handling tasks including palletizing/depalletizing and warehouse automation. However, their potential goes far beyond these conventional roles. While the manufacturing labor shortage is projected to continue and worsen, robotic capability has increased and their implementation cost has become very competitive. When a manufacturing challenge presents itself, a robotic solution is well positioned to meet or exceed project goals.

Worker shortages and high labor costs continue to be a major issue for the manufacturing sector as a whole, pushing managers to find creative ways of optimizing their understaffed labor force. As of December 2023, there are 600,000 unfilled positions within this sector according to the Bureau of Labor Statistics, and Deloitte projects an increase to 2.1 million unfilled jobs by 2030. The shortage, along with inflation, has pushed unit labor costs up 5.4% in a one year time frame according to BLS.

While the pressure to solve manufacturing challenges with reduced human labor is ongoing and pervasive, robotics is uniquely poised to offer cost-effective solutions. The cost of industrial robots has fallen substantially in the past two decades, and technological innovation has dramatically increased their capability.

Some advancements in robotics that are currently being used to creatively solve manufacturing problems include: collaborative robots, safety scanners, vision systems and large-scale data collection. Robotic implementation, programming and maintenance have become much simpler in recent years. It is worth doing the math to measure the return on investment for a robotic solution.

Key Technologies Pushing Innovative Robotics Solutions

• The Collaborative Robot: Can work directly alongside people

• Safety Scanners as a Virtual Fence: Reduces footprint, scans the whole workspace

• Trigger and Vision Systems: Part line tracking and random bin picking

• Large Scale Data Collection: Improving quality, traceability, limiting recall scope

Solutions For A Hazardous Work Environment

A recently completed project by our organization highlights the challenges faced by manufacturing industries as a whole and the potential of today’s robotics to resolve them. In this particular case, a container manufacturer wanted to explore the possibility of using a robotic solution for a dangerous repetitive task. The issue was that a worker was required to enter a poor work environment donning full personal protective equipment (PPE) gear, including a respirator, to intermittently remove high temperature containers from a conveyor for quality inspection purposes. Even more challenging, the conveyor carried several different types of parts, all requiring intermittent sampling.

Recognizing the need for a unique and novel approach, the manufacturer intentionally selected an integration partner from a different industry. They sought a partner with expertise in implementing robotic solutions in an industry with traditionally high use of automation. As an integrator serving the automotive sector and other highly automated industries, our team  was selected to first complete an engineering study and then implement the innovative robotic solution.

To solve the manufacturing problem, engineers devised a system that used several of the most recent innovations in robotics including:

Collaborative Robot: Filled with different sensing technologies, the collaborative robot (cobot) is designed to safely work near people without additional fencing requirements. Using its sensing capability, it will autonomously adjust its operating speed to maintain human safety.

A cobot was selected to work at the conveyor removing the parts. This decision was driven by the fact that beyond sampling, there continued to be a need for people to occasionally enter the workspace of the robot.

Safety Scanners as a Virtual Fence: Depending on the application, any robot can now be set up to operate guarded by a virtual fence, called a safety scanner. Because safety scanners monitor the entire work envelope of the robot, they offer more protection than a traditional light curtain, which can only watch for interruptions directly along the line of light.

Engineers designed the system to be guarded by two safety scanners. This decision supported the space constraints of the environment and also allowed ease of access to the area when needed.

Trigger and Vision Systems: Overhead light based trigger systems can be used to identify and target particular parts on a conveyor. When combined with position information from an encoder, a process called line tracking can be done, allowing the system to have knowledge of the whereabouts of a target part on the conveyor.

An overhead trigger sensor was installed to allow the PLC to identify target parts for removal from the conveyor. An encoder was mounted on the motor shaft of the conveyor, and directly connected to the cobot.

With the part in its gripper, the cobot used 2D vision capabilities to find an open slot to place the part in the inspection system. While 3D vision was not required for this solution, it offers additional capability to solve other manufacturing problems worth mentioning, including random bin picking (selection of a particular part from an unsorted parts bin) and accurate manipulation of that part for further assembly.

Using 3D vision, today’s robotics can identify, pick up and manipulate objects originating in a bin of randomly organized parts. Image courtesy of Patti Engineering.

Large-Scale Data Collection: Large amounts of manufacturing data can be analyzed for quality and reliability, process efficiency, traceability of product, limiting recall scope and root cause analysis.

By automating the sampling process, reliable inspection data for each of the molds was collected. The data provided feedback leading to process parameter adjustments which resulted in overall quality improvements.

System Operation

The system executes in the following manner:

• The conveyor transports hot units originating from several different molds. The programmable logic controller (PLC) is responsible for selecting samples from each type for inspection.
• The PLC targets a particular unit with information from a trigger sensor situated above the conveyor near the conveyor’s point of entry.
• The PLC tells the cobot to monitor that particular unit’s progress down the conveyor.
• Situated on the conveyor’s motor shaft, the encoder output is directly connected to the cobot, providing position information about the target part as it progresses along the conveyor towards the cobot.
• Using line tracking algorithms, the cobot knows when to expect the target part and pick it up.
• With the part in its gripper, the cobot then rotates to face the inspection system. It uses its own 2D vision capability to find an empty slot and place the part into it.
• The area is protected by two scanners, labeled in the diagram S1 and S2, which monitor for human motion and anything else that doesn’t belong in the cobot’s work envelope.
• A safety PLC controls the system, stopping the cobot’s operation when either of the scanners detects a person or other anomaly in its operating space.

The above diagram shows a simplified layout of the robotic inspection system described. Image courtesy of Patti Engineering.

The original operator was re-tasked to work in a safer environment and greater consistency of sampling for inspection purposes was achieved, leading to process and therefore quality improvements. Clearly, a success by all measures.

Exploring External Industry Perspectives

Projects that lack an industry standard solution can often benefit from expert knowledge from a different industry. The goal is to obtain a fresh perspective on the problem from someone with the skillset to apply their knowledge in a novel way. In fact, the automotive industry is a leading user of robotics and has a longstanding reputation of being the earliest adopter of new robotics capability. The companies and individuals who are tasked with solving automotive manufacturing challenges are uniquely equipped to bring their robotics knowledge to other industries, as was the case with the hot container problem.

About the Authors

Patti Engineering has recently published an eBook, Common Considerations Integrating Robotics in Manufacturing, that addresses these topics in further depth. Download a copy at: https://www.pattiengineering.com/blog/ebook-integrating-robotics-manufacturing/

Samuel M. Hoff is the founder and CEO of Patti Engineering. With 35 years of experience specializing in industrial automation systems, and engineering offices in Michigan, Texas, and Indianapolis, he is an industry leader.  Learn more at: https://www.pattiengineering.com/